On Wed, Aug 31, 2016 at 04:22:12PM +0200, Greg Kroah-Hartman wrote: > I can't figure out the order in which to apply your patches for this > driver, can you resend them all as a series so I know they will work > together properly? That really depends on which patches you want, i.e. which pov you take wrt. enabling the transmitter. Functions should either leave the transmitter enabled or at its previous state, but they are not consistently doing either option. Leaving the transmitter in random states is just prone to more issues. Depending on your choice, you need different patches. Transmitter unchanged: a. https://lkml.kernel.org/r/20160728083339.GA17330@laureti-dev Transmitter enabled: b. https://lkml.kernel.org/r/20160802090650.GA2946@laureti-dev The "transmitter enabled" patch touches the line containing the typo, thus the typo fix is only applicable to "transmitter unchanged". c. https://lkml.kernel.org/r/20160729075910.GA28937@laureti-dev Regardless of which route you choose, the actual race condition is fixed in: d. https://lkml.kernel.org/r/20160729074600.GA28423@laureti-dev So depending on your choice you can use: transmitter unchanged: a c d transmitter enabled: b d The patches should be independent and apply in any order (with fuzz). Thus far, Sören Brinkmann has expressed discomfort with keeping transmitter unchanged, but nobody else spoke up or commented the transmitter enabled approach. >From my pov, consistent behaviour is key as inconsistency made the race condition hard to understand and diagnose. Both routes work for that (i.e. I don't care). Before arriving at a decision, I don't think reposting patches makes sense. If all else fails, we can try consensus by lack of objections and use the "transmitter enabled" approach. Helmut -- To unsubscribe from this list: send the line "unsubscribe linux-serial" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
